There is a lot of confusion about the smart grid today. I hear all kinds of claims, along with the frequently heard “There are a lot of opinions about what the smart grid is.” Other have written that they have learned of a lot of clever little devices, but nothing that would qualify as over all smarts. The smart grid builds upon interoperability derived from the standardized telemetry and real time operations of Intelligrid. It meets the definitions of the Modern Grid Initiative. It is based upon transacted energy as described in the 2005 GridWise constitutional convention. These ideas were coded into law by the Energy Information and Security Act of 2007. There are not a lot of questions.
The problem is, the utilities are often the ones blurring things the fastest. The shining bright light of piles of federal dollars to pay for upgrades to internal problems, especially aging infrastructure, has led them to over-sell, and to over-promise. Hungry people say funny things when in sight of a free lunch.
One of the biggest barriers is a lack of architecture. The utilities have long run a seamless distribution control system, which makes a lot of sense to inexpensively manage real time events. Within that framework, direct control of devices in the end nodes makes sense as a proof of concept. But to extend beyond proof of concept, or short lived low penetration markets, a real IT architecture will need to be in place. That architecture will acknowledge interfaces, and what decision-making and capabilities should lay on either side of those interfaces.
The acknowledgement of those interfaces, and the agents that work on either side, is the moment it begins to move from “clever” to “smart” (I like that distinction). We are seeing this more in some of HAN areas, with agents built in to appliances that understand what each appliance is up to, and based on that, what sort of response is available now. In a distributed generation, or even micro-generation world, one can imagine analogous agents on the power sources.
Push this a little further by adding requirements of symmetry to each interface, and one can have device-based clever agents, responding to home (or business) based enterprise agents, negotiating with grid-based agents (whether on the micro-grid, distribution network, or at the edge of the transmission network.
To ride the clever/smart distinction a little further, intelligence in biological systems can be described as an emergent behavior of many small automata-like systems. Each of those automata faces fierce evolutionary pressures, both to do its own job better, and to interface with the whole better. If we get the interfaces right (more attention to architecture), and the security right, the smart grid can emerge from the many clever automata.
But it won’t be by applying a little Intelligrid to a substation, or by merely adding a programmable thermostat.
I am off this week to Washington where I am leading a session on the smart grid. I hope to meet some of you there.